CN114842978B - Intelligent blood gas analysis detection system and method based on medical big data - Google Patents
Intelligent blood gas analysis detection system and method based on medical big data Download PDFInfo
- Publication number
- CN114842978B CN114842978B CN202210776056.0A CN202210776056A CN114842978B CN 114842978 B CN114842978 B CN 114842978B CN 202210776056 A CN202210776056 A CN 202210776056A CN 114842978 B CN114842978 B CN 114842978B
- Authority
- CN
- China
- Prior art keywords
- acid
- blood gas
- time
- base
- real
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/145—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue
- A61B5/14503—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue invasive, e.g. introduced into the body by a catheter or needle or using implanted sensors
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/145—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue
- A61B5/14539—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue for measuring pH
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/145—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue
- A61B5/14542—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue for measuring blood gases
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
Landscapes
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Public Health (AREA)
- Medical Informatics (AREA)
- Biomedical Technology (AREA)
- Physics & Mathematics (AREA)
- Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Heart & Thoracic Surgery (AREA)
- Biophysics (AREA)
- Veterinary Medicine (AREA)
- Optics & Photonics (AREA)
- Databases & Information Systems (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
- Investigating Or Analysing Biological Materials (AREA)
Abstract
The invention relates to the technical field of big data processing, and discloses a blood gas analysis intelligent detection system and method based on medical big data, wherein a blood gas index parameter set is acquired by using a continuous blood gas monitor, real acid-base identification is carried out to obtain a real balanced pH index, and the limitation that an accurate fixed value is difficult to obtain due to built-in electrode sensing is solved; acid-base balance detection is carried out on the real balance pH index to obtain an acid-base real-time state, and balance disorder analysis is carried out by combining a blood gas index parameter set, so that the detection accuracy of double and triple disorder types is improved; and finally, an intelligent medical big data search engine system is constructed based on medical big data and an artificial intelligence technology, the acid-base imbalance type is further verified and compared, and the acid-base imbalance type is accurately detected, so that the aim of preventing misjudgment and missed judgment through intelligent detection is fulfilled, and a quick, objective and accurate judgment reference can be quickly and accurately provided for clinical treatment.
Description
Technical Field
The invention relates to the technical field of medical instruments, intelligent detection and analysis and big data processing, in particular to a blood gas analysis intelligent detection system based on medical big data.
Background
The blood gas analyzer is often applied to the medical treatment process such as acute respiratory failure diagnosis and treatment, surgical operation, rescue and monitoring, and the like, and can measure the pH value (pH) and the partial pressure of carbon dioxide (PCO) in an artery by using an electrode in a short time 2 ) And oxygen partial Pressure (PO) 2 ) And the like, and assists medical care personnel in diagnosing the acid-base balance disorder type, the respiratory compensation and other symptoms of the patient. At present, a blood gas analyzer can only accurately measure serum electrolyte and blood acid-base balance condition indexes, but most of the blood gas analyzers are sensed by built-in electrodes, the standard substance specified by regulations is complex in configuration process and short in quality guarantee period, and accurate fixed values are difficult to obtain; and the introduction of artificial uncertainty is difficult to avoid in the detection process, and medical staff is required to expend energy to calculate and judge the type of acid-base imbalance according to blood-gas parameter indexes.
The blood gas analysis and detection is a diagnosis basis for judging the type of the acid-base equilibrium disorder, and the comprehensive analysis of various indexes and the calculation and judgment process of applying the predicted compensation value are complex. In the early clinical judgment, the acid-base imbalance is judged by using an acid-base diagram method, but a large error exists, and the mixed type and three types of acid-base imbalance and other symptoms are difficult to accurately judge. At present, the diagnosis methods for judging the type of acid-base equilibrium disorder comprise a four-step method, a seven-step method, a six-step method, a staged diagnosis method for acid-base equilibrium disorder and the like, but because the calculation formulas and the steps are not uniform, the difference of conclusions obtained by different methods is large, the medical care personnel easily make erroneous judgment or missing judgment, and the accurate judgment is difficult to delay the emergency treatment.
The most prominent problem in the field of medical instruments is that high-quality medical resources are insufficient, and the accuracy and efficiency of disease diagnosis depend on the expertise and proficiency of medical care personnel to a great extent. The aging population and the high-speed increase of cardiovascular and cerebrovascular diseases in most countries and regions have the problems of serious imbalance of supply and demand of medical and human resources, uneven regional distribution and the like. Therefore, it is urgently needed to combine the artificial intelligence technology with the field of medical instruments to be applied to clinical diagnosis, so that the human resources are greatly saved, and the health requirements of people are greatly guaranteed. The rapid diagnosis and correction of the acid-base balance disorder are the key for improving the cure rate of emergency critical patients, but an intelligent detection method for the type of the acid-base balance disorder of the human body by combining medical big data is lacked at present, so that the rapid, objective and accurate judgment reference is difficult to be synchronously, rapidly and accurately provided for clinical treatment.
Disclosure of Invention
In view of the limitations of the existing methods, the present invention aims to provide an intelligent detection system and method for blood gas analysis based on medical big data, wherein the detection system comprises a blood gas index parameter preprocessing module, an acid-base real identification module, an acid-base balance detection module, a balance disorder analysis module, and a medical big data search and verification module; the continuous blood gas monitor is used for acquiring and preprocessing blood gas index parameter sets, real acid-base identification is carried out to obtain real balanced pH indexes, and the limitation that accurate fixed values are difficult to obtain due to sensing of the built-in electrodes is effectively solved; acid-base balance detection is carried out according to the real balance pH index to obtain an acid-base real-time state, balance disorder analysis is further carried out by combining with a blood gas index parameter set, the acid-base imbalance type can be obtained by covering the whole range of pH classification step-by-step accurate detection, and the detection accuracy of the double-disorder type and the triple-disorder type is improved; and finally, an intelligent medical big data search engine system is constructed based on medical big data and an artificial intelligence technology, the acid-base imbalance type is further verified and compared, and the acid-base imbalance type is accurately detected, so that the aim of intelligently detecting and preventing misjudgment and missed judgment is fulfilled, and a quick, objective and accurate judgment reference can be quickly and accurately provided for clinical treatment.
In order to achieve the above object, according to an aspect of the present disclosure, there is provided a blood gas analysis intelligent detection system based on medical big data, the system including: a processor, a memory, and a computer program stored in the memory and executed on the processor, wherein the processor executes the computer program to execute the modules comprising: the system comprises a blood gas index parameter preprocessing module, an acid-base real identification module, an acid-base balance detection module, a balance disorder analysis module and a medical big data searching and verifying module;
the blood gas index parameter preprocessing module comprises:
the blood gas index parameter acquisition unit is used for acquiring various blood gas index parameters changing along with time by using a continuous blood gas monitor;
the data preprocessing unit is used for carrying out data preprocessing on the blood gas index parameters changing along with time to obtain a blood gas index parameter set;
the acid-base real identification module is used for carrying out acid-base real identification on the blood gas index parameter set to obtain a real balanced pH index;
the acid-base balance detection module is used for carrying out acid-base balance detection on the real balance pH index to obtain an acid-base real-time state;
the balance disorder analysis module is used for carrying out balance disorder analysis on the acid-base real-time state and the blood gas index parameter set to obtain an acid-base imbalance type;
the medical big data search and verification module is used for constructing an intelligent medical big data search engine system to obtain medical big data, comparing similar parameter symptoms according to the blood gas index parameter set and the medical big data, and verifying whether the acid-base imbalance type is accurate or not;
the intelligent blood gas analysis detection system based on the medical big data runs in computing equipment of a desktop computer, a notebook computer, a palm computer and a cloud data center.
Further, the specific steps of acquiring various blood gas index parameters changing along with time by using the continuous blood gas monitor are as follows: calibrating a sensor of the continuous blood gas monitor, and then inserting the sensor into an arterial catheter to be measured to lock the sensor; placing an arterial indwelling needle for the arterial puncture to be detected to be connected with a sensor, connecting the other end of the sensor with a continuous blood gas monitor, carrying out continuous blood gas monitoring at equal intervals on the artery to be detected by using the continuous blood gas monitor, and converting each blood gas index parameter collected by the sensor into respective electric signal transmissionThe blood gas index parameters which change along with time are obtained by amplifying and analog-to-digital converting the blood gas index parameters which are transmitted to a processor of the continuous blood gas monitor; the method comprises the following specific steps of carrying out data preprocessing on various blood gas index parameters changing along with time to obtain a blood gas index parameter set: performing data conversion processing on various blood gas index parameters changing along with time, and recording to obtain a blood gas index parameter set; wherein the data conversion process is a conversion into a set format, and the set of blood gas index parameters is stored in a data set of a memory according to the time sequence that the set blood gas index parameters continuously change along with the continuous change and is recorded as BG = { pH (t), PaCO 2 (t), HCO 3 - (t), AG (t); wherein, the value of T is represented as acquisition time, and T belongs to [0, T ∈]T is total acquisition time; pH (t) is the pH at time t; PaCO 2 (t) arterial blood carbon dioxide partial pressure at time t in mmHg; HCO 3 - (t) is the actual bicarbonate concentration at time t, AG (t) is the anion gap at time t, and the formula is AG (t) = Na + (t) + Cl - (t) - HCO 3 - (t) units are mmol/L, wherein, Na + (t) is the sodium ion concentration at time t, Cl - (t) is the chloride ion concentration at time t.
Further, the method for truly identifying the acid and base of the blood gas index parameter set comprises the following specific steps of:
s301, sequentially calculating the hydrogen ion concentration index estimated value of the blood gas index parameter set at each time t to be represented as H + (t) which is calculated by the formula: h + (t)＝24×PaCO 2 (t) / HCO 3 - (t); setting a pH estimation range, further comparing whether the pH (t) corresponding to each time t is within the pH estimation range, if so, judging to mark the time t as a normal time, otherwise, marking the time t as an abnormal time;
s302, sequentially calculating the corresponding standard pH value of the blood gas index parameter set at each time t to be represented as pH1(t), wherein the calculation formula is as follows: pH1(t) = 6.1+ log [ HCO ] 3 - (t)] / PaCO 2 (t)；
S303, sequentially obtaining time intervals among all the normal moments according to the time sequence, and calculating the arithmetic mean value of all the time intervals as Mtime _ pH; acquiring the pH value corresponding to each normal moment according to the time sequence, and calculating the difference value of each pH value in sequence to be used as a pH value change sequence and recording the pH value change sequence as diff _ pH; sequentially calculating the difference value between the pH value (t) and the pH value 1(t) corresponding to each normal moment as a pH value deviation change sequence to be recorded as error _ pH;
s304, screening out elements corresponding to the arithmetic mean value of which the numerical value is less than or equal to error _ pH from diff _ pH, and sequentially storing the elements in a pH value standard variation sequence as diff _ pH 1; further screening each element of which the time interval corresponding to the time t between all adjacent elements in diff _ pH1 is less than or equal to Mtime _ pH, taking out the pH value corresponding to the element, sequentially storing the pH value corresponding to the element in a pH value standard sequence to be written as std _ pH, and calculating the arithmetic mean value of each element in std _ pH to be written as Mstd _ pH;
s305, calculating the pH value change calibration threshold of std _ pH as Th _ pH, wherein the calculation formula is as follows:
wherein max is the maximum value in the calculation sequence, min is the minimum value in the calculation sequence, std _ pH (i) is represented as the pH value corresponding to the ith element in std _ pH, i is the sequence element number, i belongs to [1, N ], and N is the total number of elements in the pH value standard sequence; further screening all elements of which the difference values between the pH values corresponding to all adjacent elements in std _ pH are less than or equal to Th _ pH, and sequentially storing the elements in a pH value real sequence to be recorded as true _ pH;
s306, calculating a true equilibrium pH index from the pH value true sequence and recording the true equilibrium pH index as Balance _ pH, wherein the calculation formula of Balance _ pH is as follows:
wherein, true _ pH (i1) is represented as the pH value corresponding to the i1 th element in true _ pH, i1 is the sequence element number, and N1 is the total number of elements of the pH value real sequence; max (s1, s2, s3) is the maximum of the calculated pH standard variables s1, s2 and s3, s1= std _ pH (i), s2= std _ pH (i +1), s3= std _ pH (i + 2).
The invention has the beneficial effects that the pH value (t) acquired by the continuous blood gas monitor can reflect the instantaneous pH value of the organism at different moments to form certain regular fluctuation change, the regular pH value change can be extracted by the blood gas index parameter preprocessing module and the acid-base real identification module, the balanced pH value within certain time can be truly reflected by further calculation through a calibration threshold value, the balanced pH value is expressed as a true balanced pH index, and the true fluctuation process of the pH value of the organism can be more accurately quantified.
Further, the specific steps of performing acid-base equilibrium detection on the true equilibrium pH index to obtain the acid-base real-time state include: when Balance _ pH is less than A1, judging the acid-base real-time state as the symptoms of acid-blood; when Balance _ pH is greater than A2, judging the acid-base real-time state as an alkaline blood symptom; otherwise, judging the real-time acid-base state to be a normal symptom; wherein A1 and A2 are acid-base threshold values, and A1=7.35 and A2=7.45 are taken.
The method comprises the steps of (a) obtaining a pH value time sequence of real-time dynamic change of the human body by utilizing a continuous blood gas monitor, (b) obtaining pH value data which accord with internal consistency in the pH value time sequence by utilizing the continuous blood gas monitor, and further carrying out acid-base real identification and calculating a real balanced pH index which is closest to the real change of the human body in continuous time by utilizing the concrete steps of the acid-base real identification module, so as to solve the problem that the built-in electrode sensing is difficult to obtain an accurate fixed value, and further carrying out acid-base balance detection by utilizing the real balanced pH index in the concrete steps of the acid-base balance detection module to obtain the acid-base real-time state under real symptoms, thereby improving the accuracy of detecting the acid-base type of the human body).
The invention also provides a blood gas analysis intelligent detection method based on medical big data, which is characterized by comprising the following steps:
s100, collecting various blood and gas index parameters changing along with time by using a continuous blood and gas monitor;
s200, carrying out data preprocessing on the blood gas index parameters changing along with time to obtain a blood gas index parameter set;
s300, performing real acid-base identification on the blood gas index parameter set to obtain a real balance pH index;
wherein the blood gas index parameter set and the true equilibrium pH index are transmitted to a memory for storage.
The invention utilizes the pH (t) acquired by the continuous blood gas monitor to reflect the instantaneous pH of the organism at different moments to form certain regular fluctuation change, can extract the regular pH change through the step S300, further calculates through a calibration threshold to obtain the balanced pH value which truly reflects the certain time and is expressed as a true balanced pH index, and realizes the true fluctuation process of the pH value of the organism more accurately.
And S400, carrying out acid-base balance detection according to the real balance pH index to obtain an acid-base real-time state.
S500, carrying out balance disorder analysis to obtain an acid-base imbalance type when the acid-base real-time state and the blood gas index parameter are combined;
s600, constructing an intelligent medical big data search engine system to obtain medical big data, and verifying whether the acid-base imbalance type is accurate or not according to the blood gas index parameter set and the medical big data.
Further, in S500, the method of analyzing the balance disorder to obtain the type of acid-base imbalance when the acid-base real-time state and the blood gas index parameter are combined includes:
s501, obtaining a blood gas index parameter set, obtaining time sequences T1 formed by corresponding moments of each element in true _ pH, and calculating all time pairs in T1The arithmetic mean of the values of PaCO2 (t) is referred to as true _ PaCO 2 (ii) a Calculating HCO corresponding to all time points in T1 3 - The arithmetic mean of the values of (t) is denoted as true _ HCO 3 - (ii) a Calculating the arithmetic mean value of the numerical values corresponding to AG (T) at all the time points in T1 and recording as true _ AG;
s502, mixing Balance _ pH and true _ PaCO 2 、true_HCO 3 - And true _ AG compares with its correspondent normal mean value separately and confirms its correspondent real-time direction; preferably, the normal average pH is 7.4, PaCO 2 Has a normal average value of 40 mmHg, HCO 3 - Has a normal average value of 24 mmol/L and a normal average value of AG is 12 mmol/L;
s503, according to the real-time states of acid and alkali, Balance _ pH and true _ PaCO are combined 2 、true_HCO 3 - And true _ AG further judging primary type and compensatory type;
s504, calculating the absolute difference between true _ AG and normal AG average as Δ AG, and calculating HCO 3 - The estimated value is represented as [ HCO ] 3 - ]The calculation formula is as follows: [ HCO ] 3 - ]=
+true_HCO 3 - (ii) a Further determining the combined acid-base imbalance type.
(in the specific steps of the acid-base balance detection module of the invention, acid-base balance detection can be performed on the true balance pH index to obtain the acid-base real-time state under the true symptom, in step S500, balance disorder analysis is performed in combination with the blood gas index parameter set, and a classification step-by-step judgment algorithm covering the pH full range is performed for the acid-base balance disorder type, so that the detection accuracy of the complex mixed double and triple acid-base balance disorder types can be improved, and the error of emergency treatment caused by the occurrence of misjudgment and misjudgment symptoms can be effectively avoided).
Further, in S600, a method for constructing an intelligent medical big data search engine system to obtain medical big data and verifying whether the acid-base imbalance type is accurate according to the blood gas index parameter set in combination with the medical big data includes:
s601, constructing an intelligent medical big data search engine system;
s602, according to the blood gas index parameter set, calculating and obtaining real blood gas analysis data through steps S200-S500 and recording the real blood gas analysis data as true _ BG, wherein true _ BG = { Balance _ pH, true _ PaCO 2 , true_HCO 3 - , true_AG}；
S603, inputting real blood gas analysis data according to the intelligent medical big data search engine system, and intelligently searching based on medical big data to obtain a standard acid-base imbalance type corresponding to a standard case with the highest similarity;
and S604, comparing the standard acid-base imbalance type with the acid-base imbalance type obtained in the step S500, if the standard acid-base imbalance type is the same as the acid-base imbalance type, verifying that the acid-base imbalance type is correct, and if the standard acid-base imbalance type is not the same as the acid-base imbalance type, verifying that the acid-base imbalance type is incorrect.
(the real blood gas analysis data reflecting the real-time acid-base state of the human body are obtained in the steps S300-S500 and used as input, an intelligent medical big data search engine system is constructed based on medical big data and an artificial intelligence technology, and the standard acid-base imbalance type corresponding to the standard case with the highest similarity can be obtained through the intelligent operation in the step S600 and compared, so that the beneficial effect of assisting diagnosis and preventing misjudgment is achieved.
Further, in S601, the method executed in the intelligent medical big data search engine system includes:
s6011, establishing an acid-base diagnosis auxiliary decision tree based on a decision tree classification algorithm of priori knowledge, and performing primary classification on standard cases through the acid-base diagnosis auxiliary decision tree to obtain blood gas analysis data corresponding to each standard case and a symptom text corresponding to the blood gas analysis data;
s6012, extracting keywords of the symptom text corresponding to each standard case respectively, performing data mining through the keywords to obtain synonyms corresponding to all the keywords, and constructing a standard symptom keyword library by the keywords and the synonyms corresponding to all the keywords;
s6013, converting each keyword in the standard symptom keyword library into a hash value based on a hash algorithm, mapping each standard case to each bit array by using a BitMap algorithm, performing ascending sorting on binary numbers of all standard cases based on bit values, and performing secondary classification through a K-modes algorithm to obtain the acid-base imbalance type.
As described above, the blood gas analysis intelligent detection system and method based on medical big data of the present invention have the following beneficial effects: (1) the pH value time sequence of real-time dynamic change of a human body can be obtained by utilizing the continuous blood gas monitor, real acid-base identification is carried out to obtain a real balanced pH index which is closest to the real change of the human body, and the limitation that an accurate fixed value is difficult to obtain due to sensing of a built-in electrode is solved; the method comprises the steps of (1) carrying out acid-base balance detection according to a real pH balance index to obtain an acid-base real-time state under real symptoms, so as to improve the accuracy of acid-base imbalance type detection, (3) carrying out classification step-by-step judgment algorithm covering the whole pH range aiming at the acid-base imbalance type to effectively prevent misjudgment and misjudgment symptoms, (4) constructing an intelligent medical big data search engine system based on medical big data and an artificial intelligence technology, and obtaining a standard acid-base imbalance type through intelligent operation to compare and verify whether the acid-base imbalance type is correct, so that the purpose of accurately and auxiliarily detecting the acid-base imbalance type is achieved, and the intelligent correction of the medical big data is realized to obtain a more accurate and efficient treatment scheme.
Drawings
The foregoing and other features of the present disclosure will become more apparent from the detailed description of the embodiments shown in conjunction with the drawings in which like reference characters designate the same or similar elements throughout the several views, and it is apparent that the drawings in the following description are merely some examples of the present disclosure and that other drawings may be derived therefrom by those skilled in the art without the benefit of any inventive faculty, and in which:
FIG. 1 is a flow chart illustrating an embodiment of a method for intelligently detecting blood gas analysis based on medical big data;
fig. 2 is a system configuration diagram of an embodiment of an intelligent blood gas analysis detection system based on medical big data.
Detailed Description
The conception, specific structure and technical effects of the present disclosure will be clearly and completely described below in conjunction with the embodiments and the accompanying drawings to fully understand the objects, aspects and effects of the present disclosure. It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict.
Fig. 1 is a flow chart of a blood gas analysis intelligent detection method based on medical big data according to the present invention, and a blood gas analysis intelligent detection method based on medical big data according to an embodiment of the present invention is described below with reference to fig. 1. The present disclosure provides a blood gas analysis intelligent detection method based on medical big data, which specifically comprises the following steps:
s100, acquiring various blood and gas index parameters changing along with time by using a continuous blood and gas monitor;
s200, carrying out data preprocessing on the blood gas index parameters changing along with time to obtain a blood gas index parameter set;
s300, carrying out real acid-base identification on the blood gas index parameter set to obtain a real balanced pH index;
and S400, carrying out acid-base balance detection according to the real balance pH index to obtain an acid-base real-time state.
S500, carrying out balance disorder analysis to obtain an acid-base imbalance type when the acid-base real-time state and the blood gas index parameter are combined;
s600, constructing an intelligent medical big data search engine system to obtain medical big data, and verifying whether the acid-base imbalance type is accurate or not according to the blood gas index parameter set and the medical big data.
Further, in S100, the method for acquiring the blood gas index parameter set by using the continuous blood gas monitor includes: calibrating a sensor of the continuous blood gas monitor, and then slowly inserting the sensor into an arterial catheter to be tested until the sensor is locked after a yellow and black line; an artery indwelling needle is placed for puncturing an artery to be detected and connected with a sensor, the other end of the sensor is connected with a continuous blood and gas monitor, the continuous blood and gas monitor is used for carrying out continuous blood and gas monitoring on the artery to be detected at equal intervals, various blood and gas index parameters collected by the sensor are converted into respective electric signals, the electric signals are transmitted to a processor of the continuous blood and gas monitor, and the blood and gas index parameters which change along with time are obtained through amplification and analog-to-digital conversion.
Preferably, the blood gas index parameters include pH value, arterial blood carbon dioxide partial pressure, bicarbonate concentration, anion gap and the like.
Further, in S200, the method for performing data preprocessing on each blood gas index parameter that changes with time to obtain a blood gas index parameter set includes: auditing and correcting the blood gas index parameters changing along with time, converting the data and recording to obtain a blood gas index parameter set; wherein the data conversion process is to convert the data into a predetermined format of the processor, and to store the predetermined blood gas index parameters into the data set of the memory according to the time sequence of the continuous change of the predetermined blood gas index parameters to obtain a blood gas index parameter set which is recorded as BG = { pH (t), PaCO = 2 (t), HCO 3 - (t), AG (t); wherein, the value of T is represented as acquisition time, and T belongs to [0, T ∈]T is total acquisition time; pH (t) is the pH at time t; PaCO 2 (t) arterial blood carbon dioxide partial pressure at time t in mmHg; HCO 3 - (t) is the actual bicarbonate concentration at time t, AG (t) is the anion gap at time t, and the formula is AG (t) = Na + (t) + Cl - (t) - HCO 3 - (t) units are mmol/L, wherein, Na + (t) is the sodium ion concentration at time t, Cl - (t) is the chloride ion concentration at time t.
Further, in S300, the method for truly identifying the acid and base of the blood gas index parameter set to obtain a true equilibrium pH index includes:
s301, sequentially calculating the hydrogen ion concentration index estimated value of the blood gas index parameter set at each time t to be represented as H + (t) the calculation formula is: h + (t)＝24×PaCO 2 (t) / HCO 3 - (t); setting the estimated pH range specifically as follows: according to pH and [ H ] + ]The corresponding pH estimation range is obtained by the comparison table with the same value, [ H ] + ]Expressed as an estimated pH value, further comparing whether the pH (t) corresponding to each time t is within the estimated pH range, if so, judging and marking the time t as a normal time; otherwise, marking the time t as abnormal time;
s302, sequentially calculating the corresponding standard pH value of the blood gas index parameter set at each time t to be represented as pH1(t), wherein the calculation formula is as follows: pH1(t) = 6.1+ log [ HCO ] 3 - (t)] / PaCO 2 (t)；
S303, sequentially obtaining time intervals among all the normal moments according to a time sequence, and calculating an arithmetic mean value of all the time intervals to be recorded as Mtime _ pH; acquiring the pH value corresponding to each normal moment according to the time sequence, and calculating the difference value of each pH value in sequence to be used as a pH value change sequence and recording the pH value change sequence as diff _ pH; sequentially calculating the difference value between the pH value (t) and the pH value 1(t) corresponding to each normal moment as a pH value deviation change sequence to be recorded as error _ pH value;
s304, screening out elements corresponding to the arithmetic mean value of which the numerical value is less than or equal to error _ pH from diff _ pH, and sequentially storing the elements in a pH value standard variation sequence as diff _ pH 1; further screening each element of which the time interval corresponding to the time t between all adjacent elements in diff _ pH1 is less than or equal to Mtime _ pH, taking out the pH value corresponding to the element, sequentially storing the pH value corresponding to the element in a pH value standard sequence to be written as std _ pH, and calculating the arithmetic mean value of each element in std _ pH to be written as Mstd _ pH;
s305, calculating the pH value change calibration threshold of std _ pH as Th _ pH, wherein the calculation formula is as follows:
wherein max is the maximum value in the calculation sequence, min is the minimum value in the calculation sequence, std _ pH (i) is represented as the pH value corresponding to the ith element in std _ pH, i is the sequence element number, i belongs to [1, N ], and N is the total number of elements in the pH value standard sequence; further screening all elements of which the difference values between the pH values corresponding to all adjacent elements in std _ pH are less than or equal to Th _ pH, and sequentially storing the elements in a pH value real sequence to be recorded as true _ pH; the difference value between the pH values corresponding to all the adjacent elements is obtained by subtracting the pH value corresponding to the current element from the pH value corresponding to the next element in the pH value standard sequence; s306, calculating a true equilibrium pH index from the pH value true sequence and recording the true equilibrium pH index as Balance _ pH, wherein the calculation formula of Balance _ pH is as follows:
wherein, true _ pH (i1) is represented as the pH value corresponding to the i1 th element in true _ pH, i1 is the sequence element number, and N1 is the total number of elements of the pH value real sequence; max (s1, s2, s3) is the maximum of the calculated pH standard variables s1, s2 and s3, s1= std _ pH (i), s2= std _ pH (i +1), s3= std _ pH (i + 2).
Specifically, the pH and [ H ] are described in S301 + ]The table of values is as follows:
further, in S400, the method for performing acid-base equilibrium detection according to the true equilibrium pH index to obtain the acid-base real-time state includes: when Balance _ pH is less than 7.35, judging the acid-base real-time state to be the symptoms of acid-blood; when Balance _ pH is greater than 7.45, judging the real-time acid-base state to be an alkaline blood symptom; otherwise, the acid-base real-time state is judged to be a normal symptom.
Wherein, the specific judgment numerical value can refer to the [1] Liu allowed construction, aging, arterial blood gas analysis and acid-base balance judgment [ J ] Chinese medical guideline 2012, 10(23):2.
Further, in S500, the method for analyzing the balance disorder by the acid-base real-time state and the blood gas index parameter set to obtain the acid-base imbalance type includes:
s501, obtaining a blood gas index parameter set, obtaining time sequences T1 formed by corresponding time of each element in true _ pH, calculating an arithmetic mean of values of PaCO2 (T) corresponding to all time in T1 and recording the arithmetic mean as true _ PaCO 2 (ii) a Calculating HCO corresponding to all time points in T1 3 - The arithmetic mean of the values of (t) is denoted as true _ HCO 3 - (ii) a Calculating the arithmetic mean value of the numerical values corresponding to AG (T) at all the time points in T1 and recording as true _ AG;
s502, mixing Balance _ pH and true _ PaCO 2 、true_HCO 3 - And true _ AG compares with its correspondent normal mean value separately and confirms its correspondent real-time direction;
preferably, in S502, the method for comparing and determining the corresponding real-time direction includes: the comparison is large, namely the real-time direction is enlarged; the comparison is smaller, namely the real-time direction is smaller;
wherein the corresponding normal average values include: normal pH average 7.4, normal PaCO 2 Average value of 40 mmHg, normal HCO 3 - The average value is 24 mmol/L, and the average value of normal AG is 12 mmol/L;
preferably, in S502, the real-time direction of Balance _ pH is determined, specifically: comparing the Balance _ pH value with the normal pH average value, wherein the real-time direction is increased when the Balance _ pH value is larger, and the real-time direction is decreased when the Balance _ pH value is larger; determining true _ PaCO 2 Real-time direction of (2): will true _ PaCO 2 And normal PaCO 2 Comparison of the mean values, true _ PaCO 2 At larger time, i.e. the real-time direction becomes larger, true _ PaCO 2 The real-time direction becomes smaller when the direction is larger; determining true _ HCO 3 - Real-time direction of (2): will true _ HCO 3 - With normal HCO 3 - The average value is compared, when the value of true _ HCO3 is larger, namely the real-time direction is larger, true _ HCO 3 - The real-time direction becomes smaller when the direction is larger; determining the real-time direction of true _ AG: comparing the true _ AG with the normal AG average value, wherein the real-time direction is increased when the true _ AG is larger, and the real-time direction is decreased when the true _ AG is larger; wherein the normal pH average is 7.4, normal PaCO 2 Average value of 40 mmHg, normal HCO 3 - The average value is 24 mmol/L, and the average value of normal AG is 12 mmol/L;
s503, according to the real-time states of acid and alkali, Balance _ pH and true _ PaCO are combined 2 、true_HCO 3 - And true _ AG further determines primaryA sex type and a compensatory type;
preferably, in S503, when the real-time acid-base status is not a normal symptom, the method for further judging the primary type is:
s5031, judging Balance _ pH and true _ PaCO when the real-time acid-base state is alkaline blood symptom 2 Whether the real-time directions of (a) are consistent: when the real-time directions are consistent, determining the direction as primary metabolic disorder; when the real-time directions are inconsistent, determining the direction as primary respiratory obstruction;
s5032, further judging the primary type of acidosis symptoms: when the disease is primary metabolic disorder and the Balance _ pH real-time direction is changed to be small, determining the disease as primary metabolic acidosis symptom; when the disease is primary metabolic disorder and the Balance _ pH real-time direction is increased, determining the disease as primary metabolic alkalosis symptom; when the respiratory disorder is primary respiratory disorder and the Balance _ pH real-time direction is smaller, determining the respiratory disorder as primary respiratory acidosis symptom; when the disease is primary respiratory disturbance and the Balance _ pH real-time direction is increased, determining the disease as primary respiratory alkalosis symptom;
specifically, the method for further determining the primary type in S5031 to S5032 can be represented by the following table:
here, ↓ indicates a real-time direction becoming smaller, and ↓ indicates a real-time direction becoming larger.
Preferably, in S503, when the real-time acid-base status is a normal symptom, the method for further judging the type of compensation is:
s5033, judging the Balance as partial acidity compensation when Balance _ pH is less than or equal to 7.35 and less than 7.40;
s5034, judging the alkaline compensation when the pH value in the Balance _ is less than 7.40 and less than or equal to 7.45;
s5035, judged as normal when Balance _ pH = 7.40; further judging the type of the acid substitute: when true _ AG real-time direction is large and true _ HCO 3 - Direction in real time is diminished, Cl - Determining AG-increased type acid substitutes in normal conditions; when the true _ AG is normal,true_ HCO 3 - real time direction is diminished and Cl - When the increase is increased, determining the AG normal type acid substitute; when the real-time direction of true _ AG is larger, true _ HCO 3 - Real time direction is diminished and Cl - When the height is increased, the mixed acid is determined; when the real-time direction of true _ AG is changed to be small, determining the blood type of the low protein blood type;
s5036, calculating a predicted compensation range through a predicted compensation formula, and further determining the type of the simple acid-base imbalance;
preferably, after determining the primary type or the compensatory type in step S503, the compensatory property is further evaluated, and the predicted compensatory range is calculated by the predicted compensatory formula in step S5036 to determine the simple acid-base imbalance type:
(1) the compensatory properties were evaluated on the basis of the primary metabolic acidosis symptoms: by the formula of acid substitution compensation: PaCO 2 = 1.5 HCO 3 - +8 +/-2 calculation, judging true _ PaCO 2 Whether it is within the expected compensation range: if so, determining the simple primary metabolic acidosis; otherwise, if true _ PaCO 2 Beyond the predicted compensation range, determining as primary metabolic acidosis + respiratory acidosis; if true _ PaCO 2 When the concentration is lower than the expected compensation range, the concentration is determined as primary metabolic acidosis + respiratory alkalosis;
(2) the compensatory properties were evaluated from primary respiratory acidosis:
A. the formula of acute acid compensation is as follows: HCO 3 - =24+(PaCO 2 -40) x 0.07 + -1.5, judge true _ HCO 3 - Whether it is within the expected compensation range: if yes, determining the patient is simple primary respiratory acidosis; otherwise, if true _ HCO 3 - Beyond the predicted compensation range, determining the primary respiratory acidosis + metabolic alkalosis; if true _ HCO 3 - When the concentration is lower than the expected compensation range, the concentration is determined as primary respiratory acidosis + metabolic acidosis;
B. the formula is compensated by chronic respiratory acid: HCO 3 - =24 +(PaCO 2 -40) x 0.4 + -3, judge true _ HCO 3 - Whether or not within the expected compensation rangeThe method comprises the following steps: if yes, determining the patient is simple primary respiratory acidosis; otherwise, if true _ HCO 3 - Beyond the predicted compensation range, determining the primary respiratory acidosis + metabolic alkalosis; if true _ HCO 3 - When the concentration is lower than the expected compensation range, the concentration is determined as primary respiratory acidosis + metabolic acidosis;
(3) the compensatory properties were evaluated from primary metabolic alkalosis: by the compensation formula: PaCO 2 =40+(HCO 3 - -24) x 0.9 ± 5, judge true _ PaCO 2 Whether it is within the expected compensation range: if yes, determining the primary metabolic alkalosis as simple primary metabolic alkalosis; otherwise, if true _ PaCO 2 Beyond the predicted compensation range, determining the primary metabolic alkalosis + respiratory acidosis; if true _ PaCO 2 When the concentration is lower than the expected compensation range, determining the concentration as primary metabolic alkalosis + respiratory alkalosis;
(4) the compensatory properties were evaluated according to primary respiratory alkalosis:
A. the acute respiratory alkalosis compensation formula is as follows: HCO 3 - =24-(40- PaCO 2 ) Calculating x 0.2 +/-2.5, and judging true _ HCO 3 - Whether it is within the expected compensation range: if yes, determining the simple primary respiratory alkalosis; otherwise, if true _ HCO 3 - Beyond the predicted compensation range, determining the primary respiratory alkalosis + metabolic alkalosis; if true _ HCO 3 - When the concentration is lower than the expected compensation range, the concentration is determined as primary respiratory alkalosis + metabolic acidosis;
B. the formula for compensating for chronic respiratory alkalosis is as follows: HCO 3 - =24-(40- PaCO 2 ) Calculating x 0.5 +/-2.5, and judging true _ HCO 3 - Whether it is within the expected compensation range: if yes, determining the simple primary respiratory alkalosis; otherwise, if true _ HCO 3 - Beyond the predicted compensation range, determining the primary respiratory alkalosis + metabolic alkalosis; if true _ HCO 3 - And when the concentration is lower than the expected compensation range, determining the concentration as primary respiratory alkalosis + metabolic acidosis.
Specifically, the predicted compensation formula in S5036 can be shown as follows:
s504, calculating the absolute difference between true _ AG and normal AG average value as Δ AG, and calculating HCO 3 - The estimated value is expressed as [ HCO 3 - ]The calculation formula is as follows: [ HCO ] 3 - ]=△AG+true_HCO 3 - (ii) a Further determining the combined acid-base imbalance type.
Preferably, the method for determining the combined acid-base imbalance type in S504 is as follows: if [ HCO ] 3 - ]<22, determined to be combined with metabolic acidosis; if [ HCO ] 3 - ]>26, identified as combined metabolic alkalosis; if 22 is less than or equal to [ HCO ] 3 - ]Less than or equal to 26, and is determined as simple acid-base equilibrium disorder.
In particular, reference may be made to the primary and compensatory types:
[1] acid-base balance and blood-gas analysis of Chen Xuanming [ J ] occupational and health, 2000, 16(1):3.
[2] Compensation and primary change of plasma [ HCO _ 3- ] when Liujing is derived and respiratory acid (alkali) is poisoned [ J ]. China modern medicine journal, 2005.
Further, in S600, an intelligent medical big data search engine system is constructed to obtain medical big data, and the method for verifying whether the acid-base imbalance type is correct according to the blood gas index parameter set and the medical big data includes:
s601, constructing an intelligent medical big data search engine system, wherein the medical big data are derived from pre-stored standard cases;
s602, according to the blood gas index parameter set, calculating and obtaining real blood gas analysis data through steps S200-S500 and recording the real blood gas analysis data as true _ BG, wherein true _ BG = { Balance _ pH, true _ PaCO 2 , true_HCO 3 - , true_AG}；
S603, inputting real blood gas analysis data according to the intelligent medical big data search engine system, and intelligently searching based on medical big data to obtain a standard acid-base imbalance type corresponding to a standard case with the highest similarity;
s604, comparing the standard acid-base imbalance type with the acid-base imbalance type obtained in the step S500, if the standard acid-base imbalance type is the same, checking that the acid-base imbalance type is correct, and if the standard acid-base imbalance type is not the same, checking that the acid-base imbalance type is incorrect.
Further, in S601, the method for constructing the intelligent medical big data search engine system includes:
s6011, establishing an acid-base diagnosis auxiliary decision tree based on a decision tree classification algorithm of priori knowledge, and performing primary classification on standard cases through the acid-base diagnosis auxiliary decision tree to obtain blood gas analysis data corresponding to each standard case and a symptom text corresponding to the blood gas analysis data;
s6012, extracting keywords of the symptom texts corresponding to the standard cases respectively, mining synonyms corresponding to all the keywords by means of a search engine word library, and constructing a standard symptom keyword library; preferably, the keywords of the texts of the standard cases corresponding to various acid-base imbalance types are extracted by an inverse maximum matching method; extracting key words of blood gas index parameters corresponding to various acid-base imbalance types based on one or a combination of TF-IDF algorithm, Topic-model algorithm, TextRank algorithm and the like; the search engine word stock can be one of hundred-degree search, Google search and the like;
s6013, converting each keyword in the standard symptom keyword library into a hash value based on a hash algorithm, mapping each standard case to each bit array by using a BitMap algorithm, performing ascending sequencing on binary numbers of all standard cases based on bit values, and performing secondary classification through a K-models algorithm to obtain an acid-base imbalance type;
and S6014, obtaining S6011-S6013 as an intelligent medical big data search engine system, and outputting to obtain a corresponding acid-base imbalance type by inputting corresponding blood gas analysis data for searching.
Fig. 2 is a structural diagram of a blood gas analysis intelligent detection system based on medical big data according to the present disclosure, and the processor executes the computer program to run in the following modules of the system: the system comprises a blood gas index parameter preprocessing module, an acid-base real identification module, an acid-base balance detection module, a balance disorder analysis module and a medical big data search and verification module; wherein, the first and the second end of the pipe are connected with each other,
the blood gas index parameter preprocessing module comprises:
the blood gas index parameter acquisition unit is used for acquiring various blood gas index parameters changing along with time by using a continuous blood gas monitor;
the data preprocessing unit is used for carrying out data preprocessing on the blood gas index parameters changing along with time to obtain a blood gas index parameter set;
the acid-base real identification module is used for carrying out acid-base real identification on the blood gas index parameter set to obtain a real balanced pH index;
the acid-base balance detection module is used for carrying out acid-base balance detection on the real balance pH index to obtain an acid-base real-time state;
the balance disorder analysis module is used for carrying out balance disorder analysis on the acid-base real-time state and the blood gas index parameter set to obtain an acid-base imbalance type;
the medical big data search and verification module is used for constructing an intelligent medical big data search engine system to obtain medical big data, comparing similar parameter symptoms according to the blood gas index parameter set and the medical big data, and verifying whether the acid-base imbalance type is accurate or not;
the intelligent blood gas analysis detection system based on the medical big data can be operated in computing equipment such as a desktop computer, a notebook computer, a palm computer and a cloud data center. The blood gas analysis intelligent detection system based on medical big data comprises, but is not limited to, a processor and a memory. It will be understood by those skilled in the art that the example is only an example of the blood gas analysis intelligent detection method and system based on the medical big data, and does not constitute a limitation to the blood gas analysis intelligent detection method and system based on the medical big data, and may include more or less components than the medical big data, or combine some components, or different components, for example, the blood gas analysis intelligent detection system based on the medical big data may further include an input and output device, a network access device, a bus, and the like.
The Processor may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete component Gate or transistor logic device, discrete hardware component, etc. The general processor can be a microprocessor or the processor can be any conventional processor, and the processor is a control center of the blood gas analysis intelligent detection system based on the medical big data, and various interfaces and lines are utilized to connect various subareas of the whole blood gas analysis intelligent detection system based on the medical big data.
The memory can be used for storing the computer program and/or the module, and the processor realizes various functions of the intelligent blood gas analysis detection method and system based on medical big data by running or executing the computer program and/or the module stored in the memory and calling the data stored in the memory. The memory may primarily include a program storage area and a data storage area, wherein the memory may include a high speed random access memory, and may further include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), at least one magnetic disk storage device, a Flash memory device, or other volatile solid state storage device.
Preferably, all undefined variables in the present invention may be threshold values set manually if they are not defined explicitly.
Although the description of the present disclosure has been rather exhaustive and particularly described with respect to several illustrated embodiments, it is not intended to be limited to any such details or embodiments or any particular embodiments, so as to effectively encompass the intended scope of the present disclosure. Furthermore, the foregoing describes the disclosure in terms of embodiments foreseen by the inventor for which an enabling description was available, notwithstanding that insubstantial modifications of the disclosure, not presently foreseen, may nonetheless represent equivalent modifications thereto.
Claims (7)
1. A blood gas analysis intelligent detection system based on medical big data is characterized in that the system comprises: a processor, a memory, and a computer program stored in the memory and executed on the processor, wherein the processor executes the computer program to execute the modules comprising: the system comprises a blood gas index parameter preprocessing module, an acid-base real identification module, an acid-base balance detection module, a balance disorder analysis module and a medical big data searching and verifying module;
the blood gas index parameter preprocessing module comprises:
the blood gas index parameter acquisition unit is used for acquiring various blood gas index parameters changing along with time by using a continuous blood gas monitor;
the data preprocessing unit is used for carrying out data preprocessing on the blood gas index parameters changing along with time to obtain a blood gas index parameter set;
the acid-base real identification module is used for carrying out acid-base real identification on the blood gas index parameter set to obtain a real balanced pH index;
in the acid-base real identification module, the acid-base real identification is carried out on the blood gas index parameter set, and the specific steps of obtaining the real balanced pH index comprise:
s301, sequentially calculating the hydrogen ion concentration index estimated value of the blood gas index parameter set at each time t to be represented as H + (t) the calculation formula is: h + (t)＝24×PaCO 2 (t) / HCO 3 - (t); setting a pH estimation range, further comparing whether the pH (t) corresponding to each time t is within the pH estimation range, if so, judging to mark the time t as a normal time, otherwise, marking the time t as an abnormal time;
s302, sequentially calculating the corresponding standard pH value of the blood gas index parameter set at each time t to be represented as pH1(t), wherein the calculation formula is as follows: pH1(t) = 6.1+ log 10 [HCO 3 - (t)]/PaCO 2 (t)；
S303, sequentially obtaining time intervals among all the normal moments according to a time sequence, and calculating an arithmetic mean value of all the time intervals to be recorded as Mtime _ pH; acquiring the pH value corresponding to each normal moment according to the time sequence, and calculating the difference value of each pH value in sequence to be used as a pH value change sequence and recording the pH value change sequence as diff _ pH; sequentially calculating the difference value between the pH value (t) and the pH value 1(t) corresponding to each normal moment as a pH value deviation change sequence to be recorded as error _ pH;
s304, screening out elements corresponding to the arithmetic mean value of which the numerical value is less than or equal to error _ pH from diff _ pH, and sequentially storing the elements in a pH value standard variation sequence as diff _ pH 1; further screening each element of which the time interval corresponding to the time t between all adjacent elements in diff _ pH1 is less than or equal to Mtime _ pH, taking out the pH value corresponding to the element, sequentially storing the pH value corresponding to the element in a pH value standard sequence to be written as std _ pH, and calculating the arithmetic mean value of each element in std _ pH to be written as Mstd _ pH;
s305, calculating the pH value change calibration threshold of std _ pH as Th _ pH, wherein the calculation formula is as follows:
wherein max is the maximum value in the calculation sequence, min is the minimum value in the calculation sequence, std _ pH (i) is represented as the pH value corresponding to the ith element in std _ pH, i is the sequence element number, i belongs to [1, N ], and N is the total number of elements in the pH value standard sequence; further screening all elements of which the difference values between the pH values corresponding to all adjacent elements in std _ pH are less than or equal to Th _ pH, and sequentially storing the elements in a pH value real sequence to be recorded as true _ pH;
s306, calculating a true equilibrium pH index from the pH value true sequence and recording the true equilibrium pH index as Balance _ pH, wherein the calculation formula of Balance _ pH is as follows:
wherein, true _ pH (i1) is represented as the pH value corresponding to the i1 th element in true _ pH, i1 is the sequence element number, and N1 is the total number of elements of the pH value real sequence; max (s1, s2, s3) is the maximum of the calculated pH standard variables s1, s2 and s3, s1= std _ pH (i), s2= std _ pH (i +1), s3= std _ pH (i + 2);
the acid-base balance detection module is used for carrying out acid-base balance detection on the real balance pH index to obtain an acid-base real-time state;
the balance disorder analysis module is used for carrying out balance disorder analysis on the acid-base real-time state and the blood gas index parameter set to obtain an acid-base imbalance type;
the medical big data search and verification module is used for constructing an intelligent medical big data search engine system to obtain medical big data, comparing similar parameter symptoms according to the blood gas index parameter set and the medical big data, and verifying whether the acid-base imbalance type is accurate or not;
in the medical big data search and verification module, an intelligent medical big data search engine system is constructed to obtain medical big data, similar parameter symptom comparison is carried out according to the blood gas index parameter set and the medical big data, and the specific method for verifying whether the acid-base imbalance type is accurate is as follows:
s601, constructing an intelligent medical big data search engine system;
s602, according to the blood gas index parameter set, real blood gas analysis data obtained through calculation by the data preprocessing unit, the acid-base real identification module, the acid-base Balance detection module and the Balance disorder analysis module is recorded as true _ BG, wherein true _ BG = { Balance _ pH, true _ PaCO = 2 , true_HCO 3 - , true_AG}；
S603, inputting real blood gas analysis data according to the intelligent medical big data search engine system, and intelligently searching based on medical big data to obtain a standard acid-base imbalance type corresponding to a standard case with the highest similarity;
s604, comparing the standard acid-base imbalance type with the acid-base imbalance type obtained by the imbalance analysis module, if the standard acid-base imbalance type is the same as the acid-base imbalance type, verifying that the acid-base imbalance type is correct, and if the standard acid-base imbalance type is not the same as the acid-base imbalance type, verifying that the acid-base imbalance type is incorrect;
the system runs in computing equipment of a desktop computer, a notebook computer, a palm computer and a cloud data center.
2. The medical big data-based blood gas analysis intelligent detection system as claimed in claim 1, wherein the specific steps of acquiring various time-varying blood gas index parameters by using the continuous blood gas monitor are as follows: calibrating a sensor of the continuous blood gas monitor, and then inserting the sensor into an arterial catheter to be measured to lock the sensor; placing an arterial indwelling needle for puncturing an artery to be detected to be connected with a sensor, connecting the other end of the sensor with a continuous blood gas monitor, carrying out continuous blood gas monitoring at equal intervals on the artery to be detected by using the continuous blood gas monitor, converting various blood gas index parameters collected by the sensor into respective electric signals, transmitting the respective electric signals to a processor of the continuous blood gas monitor, and obtaining various blood gas index parameters changing along with time through amplification and analog-to-digital conversion; the method comprises the following specific steps of carrying out data preprocessing on various blood gas index parameters changing along with time to obtain a blood gas index parameter set: performing data conversion processing on various blood gas index parameters changing along with time, and recording to obtain a blood gas index parameter set; wherein the data conversion process is a conversion into a set format, and the set of blood gas index parameters is stored in a data set of a memory according to the time sequence that the set blood gas index parameters continuously change along with the continuous change and is recorded as BG = { pH (t), PaCO 2 (t), HCO 3 - (t), AG (t); wherein, the value of T is represented as acquisition time, and T belongs to [0, T ∈]T is total acquisition time; pH (t) is the pH at time t; PaCO 2 (t) arterial blood carbon dioxide partial pressure at time t in mmHg; HCO 3 - (t) is the actual bicarbonate concentration at time t, AG (t) is the anion gap at time t, and the formula is AG (t) = Na + (t) + Cl - (t) - HCO 3 - (t) units are mmol/L, wherein, Na + (t) is the sodium ion concentration at time t, Cl - (t) is the chloride ion concentration at time t.
3. The intelligent detection system for blood gas analysis based on medical big data as claimed in claim 1, wherein the specific steps of performing acid-base balance detection on the true balance pH index to obtain the real-time acid-base state comprise: when Balance _ pH is less than A1, judging the acid-base real-time state as the symptoms of acid-blood; when Balance _ pH is greater than A2, judging the acid-base real-time state as an alkaline blood symptom; otherwise, judging the real-time acid-base state to be a normal symptom; wherein A1 and A2 are acid-base threshold values.
4. A blood gas analysis smart detection method of a smart detection system according to any one of claims 1-3, wherein the method comprises the steps of:
s100, collecting various blood and gas index parameters changing along with time by using a continuous blood and gas monitor;
s200, carrying out data preprocessing on the blood gas index parameters changing along with time to obtain a blood gas index parameter set;
s300, performing real acid-base identification on the blood gas index parameter set to obtain a real balance pH index;
wherein, the blood gas index parameter set and the true equilibrium pH index are transmitted to a memory for storage.
5. The intelligent blood gas analysis detection method of an intelligent detection system according to claim 4, further comprising: s400, carrying out acid-base balance detection on the real balance pH index to obtain an acid-base real-time state, wherein the method specifically comprises the following steps: when Balance _ pH is less than A1, judging the acid-base real-time state as the symptoms of acid-blood; when Balance _ pH is greater than A2, judging the acid-base real-time state as an alkaline blood symptom; otherwise, judging the real-time acid-base state to be a normal symptom; wherein A1 and A2 are acid-base threshold values.
6. The intelligent blood gas analysis detection method of an intelligent detection system according to claim 5, further comprising: s500, carrying out balance disorder analysis to obtain the acid-base imbalance type when the acid-base real-time state and the blood gas index parameter are combined, wherein the specific method comprises the following steps:
s501, obtaining a blood gas index parameter set, obtaining time sequences T1 formed by corresponding time of each element in true _ pH, calculating an arithmetic mean of values of PaCO2 (T) corresponding to all time in T1 and recording the arithmetic mean as true _ PaCO 2 (ii) a Calculating HCO corresponding to all time points in T1 3 - The arithmetic mean of the values of (t) is denoted as true _ HCO 3 - (ii) a Calculating the arithmetic mean value of the numerical values corresponding to AG (T) at all the time points in T1 and recording as true _ AG;
s502, mixing Balance _ pH and true _ PaCO 2 、true_HCO 3 - And the true _ AG is respectively compared with the corresponding normal average value to determine the corresponding real-time direction;
s503, according to the real-time states of acid and alkali, Balance _ pH and true _ PaCO are combined 2 、true_HCO 3 - And true _ AG further judging primary type and compensatory type;
s504, calculating the absolute difference between true _ AG and normal AG average value as Δ AG, and calculating HCO 3 - The estimated value is represented as [ HCO ] 3 - ]The calculation formula is as follows: [ HCO ] 3 - ]=△AG+true_HCO 3 - (ii) a Further determining the combined acid-base imbalance type.
7. The blood gas analysis intelligent detection method of the intelligent detection system according to claim 6, wherein in S601, the method executed in the intelligent medical big data search engine system is constructed to include:
s6011, establishing an acid-base diagnosis assistant decision tree based on a decision tree classification algorithm of priori knowledge, and performing primary classification on standard cases through the acid-base diagnosis assistant decision tree to obtain blood gas analysis data corresponding to each standard case and symptom texts corresponding to the blood gas analysis data;
s6012, extracting keywords of the symptom text corresponding to each standard case respectively, performing data mining through the keywords to obtain synonyms corresponding to all the keywords, and constructing a standard symptom keyword library by the keywords and the synonyms corresponding to all the keywords;
s6013, converting each keyword in the standard symptom keyword library into a hash value based on a hash algorithm, mapping each standard case to each bit array by using a BitMap algorithm, sequencing all the standard cases in an ascending order based on binary numbers of bit values, and performing secondary classification through a K-modes algorithm to obtain the acid-base imbalance type.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210776056.0A CN114842978B (en) | 2022-07-04 | 2022-07-04 | Intelligent blood gas analysis detection system and method based on medical big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210776056.0A CN114842978B (en) | 2022-07-04 | 2022-07-04 | Intelligent blood gas analysis detection system and method based on medical big data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114842978A CN114842978A (en) | 2022-08-02 |
| CN114842978B true CN114842978B (en) | 2022-09-27 |
Family
ID=82573501
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210776056.0A Active CN114842978B (en) | 2022-07-04 | 2022-07-04 | Intelligent blood gas analysis detection system and method based on medical big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114842978B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112071416A (en) * | 2020-09-01 | 2020-12-11 | 新疆维吾尔自治区人民医院 | Blood gas analysis and judgment simulation system and method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101533008A (en) * | 2009-04-23 | 2009-09-16 | 杨亚东 | Method for judging human body blood-qi acid-base disorder type |
| WO2010027339A2 (en) * | 2008-09-08 | 2010-03-11 | Kovac Anton | The method of electronic transcutanne analysis of inner blood surrounding parameters |
| CN106021968A (en) * | 2016-07-13 | 2016-10-12 | 佛山科学技术学院 | Clinical intelligent blood gas analysis and diagnosis instrument |
| CN107610782A (en) * | 2017-08-28 | 2018-01-19 | 佛山科学技术学院 | A kind of acid-base balance state analysis method and system |
| CN109009155A (en) * | 2018-08-23 | 2018-12-18 | 广州医科大学附属第二医院 | A kind of Blood_gas analysis method |
| CN111177356A (en) * | 2019-12-31 | 2020-05-19 | 佛山科学技术学院 | Acid-base index medical big data analysis method and system |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109239322A (en) * | 2018-09-05 | 2019-01-18 | 芜湖森爱驰生物科技有限公司 | A kind of intelligent blood gas analysis system and method for wireless transmission |
| EP3721804A1 (en) * | 2019-04-11 | 2020-10-14 | Obi ApS | Identification and quantification of a ventilatory distubance causing incorrect measurement of arterial acid-base status |
| CN112002412A (en) * | 2020-08-23 | 2020-11-27 | 吾征智能技术(北京)有限公司 | System, equipment and storage medium for deducing disease based on blood routine data |
-
2022
- 2022-07-04 CN CN202210776056.0A patent/CN114842978B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010027339A2 (en) * | 2008-09-08 | 2010-03-11 | Kovac Anton | The method of electronic transcutanne analysis of inner blood surrounding parameters |
| CN101533008A (en) * | 2009-04-23 | 2009-09-16 | 杨亚东 | Method for judging human body blood-qi acid-base disorder type |
| CN106021968A (en) * | 2016-07-13 | 2016-10-12 | 佛山科学技术学院 | Clinical intelligent blood gas analysis and diagnosis instrument |
| CN107610782A (en) * | 2017-08-28 | 2018-01-19 | 佛山科学技术学院 | A kind of acid-base balance state analysis method and system |
| CN109009155A (en) * | 2018-08-23 | 2018-12-18 | 广州医科大学附属第二医院 | A kind of Blood_gas analysis method |
| CN111177356A (en) * | 2019-12-31 | 2020-05-19 | 佛山科学技术学院 | Acid-base index medical big data analysis method and system |
Non-Patent Citations (3)
| Title |
|---|
| 动脉血气分析和治疗建议软件的开发和应用;李凯述等;《医学信息》;20061231;第19卷(第12期);第2084-2087页 * |
| 基于Web的血气酸碱失衡计算机辅助分析系统的研发;王银改等;《中国医疗设备》;20111231;第26卷(第12期);第34-37页 * |
| 开发自动判断酸碱失衡类型的计算机软件系统及应用;胡茂胜等;《中国现代医学杂志》;20060331;第16卷(第05期);第759-761页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114842978A (en) | 2022-08-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112669960B (en) | Method for constructing liver fibrosis prediction model based on machine learning method, prediction system, equipment and storage medium | |
| CN114842978B (en) | Intelligent blood gas analysis detection system and method based on medical big data | |
| CN111274400B (en) | Construction method, device, equipment and storage medium of medical term system | |
| JP2002022748A (en) | Clinical inspection system | |
| CN115083604B (en) | Method, system and prediction device for constructing postoperative adverse event prediction model | |
| US20220246302A1 (en) | Data analysis apparatus, data analysis method, and data analysis program | |
| CN110379503B (en) | Online fault detection and diagnosis system based on continuous blood glucose monitoring system | |
| CN110448306B (en) | Online fault detection and diagnosis method based on continuous blood glucose monitoring system | |
| CN112183572A (en) | Method and device for generating prediction model for predicting pneumonia severity | |
| JP2012505456A (en) | Improvements in multi-parameter monitoring or improvements related to multi-parameter monitoring | |
| CN111128327A (en) | Hypoglycemia early warning method and device | |
| Sauki et al. | Assessing the Asynchrony Event Based on the Ventilation Mode for Mechanically Ventilated Patients in ICU | |
| US20220351382A1 (en) | Non-Invasive Method Of Evaluating Blood Cell Measurement And Non-Invasive Blood Cell Measurement Evaluating System | |
| CN116130104A (en) | ICU patient blood glucose prediction model construction method based on machine learning | |
| Sang et al. | Study on survival prediction of patients with heart failure based on support vector machine algorithm | |
| CN113822420A (en) | Dead space fraction model building method and system based on volumetric carbon dioxide map | |
| Berdaly et al. | Comparative machine-learning approach: study for heart diseases | |
| Melillo et al. | Heart rate variability for automatic assessment of congestive heart failure severity | |
| CN112071416A (en) | Blood gas analysis and judgment simulation system and method | |
| CN116153516B (en) | Disease big data mining analysis system based on distributed computing | |
| US20220338741A1 (en) | Detection of reliable blood pressure measurements | |
| WO2019096869A1 (en) | Method for detecting abnormal values of a biomarker | |
| TWI834520B (en) | Method for establishing a model for analyzing the correlation degree of the white robe effect and its computing device, and a method for analyzing the correlation degree of the white robe effect and its computing device | |
| Sun et al. | An optimal decision tree model for diabetes diagnosis | |
| CN114548259B (en) | PISA fault identification method based on Semi-supervised Semi-KNN model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |